While IIIF has focused primarily on Image based content, there has always been interest around the community in extending the paradigm to include Video and Audio resources, following the same successful pattern as for Images. Early experiments in this direction included the linking of audio segments to the regions of a canvas that depict a musical score and linking video of commentary about the resource. More recently, the British Library was successful in obtaining a grant to work on preserving and making accessible audio content, with explicit resourcing for advancing IIIF's specifications in this area.
This group will create technical specifications that improve the interoperability of access to audio and video content following the same patterns as the Image API, and allow the integration of that content with the Presentation API. It will assist with and steer the implementation of community infrastructure, such as reference implementations and validators, as well as transformation tools to generate the required data from existing systems and APIs.
If successful, the work will enable the same degree of interoperability for audio and video resources as the community has already provided for image based resources. This will create greater access to our digital and digitized time-based cultural heritage.
The scope of the group's efforts are divided into two primary areas, access to the bitstreams of the audio and video content, and integration with and extension of the existing Presentation API to accommodate its appropriate rendering.
Work that is out of scope for this group includes the creation of client-side integration tooling for existing third party APIs, such as controlling site-specific videos in the same way as content provided more HTML5-friendly content. While useful, this would be dependent on the API version of the third party systems, over which we have no control or method of providing feedback. This is more suitable for the "shimming" process or re-hosting of the content.
The group commits to following the requirements for the specification process, including the production of two independent implementations of each feature specified, and to reach out to other communities for feedback and to encourage adoption.
The Audio and Video APIs will mirror the existing Image API in function - they provide interoperable access to temporal and spatial segmentation and transformation of the content to enable its use and reuse in multiple contexts. This has the same division as the Image API between access to the bits and a JSON-LD description of the features provided by the service.
The functionality to be provided by the APIs was scoped at a workshop in April 2016, to be confirmed and strengthened by additional use cases and implementation experience. The interactions are to focus on web-based delivery of the content, but not to the exclusion of non-browser-based client systems.
Anticipated Deliverables:
- Specification for Audio API that provides access to audio bitstreams
- Specification for Video API that provides access to video bitstreams
- Reference implementations of those APIs
- Validation services for those APIs
The Presentation API and its underlying model, Shared Canvas, will be updated to include the missing facilities for time-based media. In particular, the following significant changes are anticipated:
- Add a duration dimension to Canvases, and make height/width non-mandatory
- ViewingHints for transitions between canvases
- Recast the Annotation section as a separate cookbook, and define a/v annotation patterns such as closed captions
- Update the Annotation model to W3C standard, and adopt further community conventions
Anticipated Deliverables:
- Version 3.0 of the Presentation API, extended to enable the rendering of time-based media
- Reference implementations of the API
- Updated validation services for the new version
- Q2 2016: Initial workshop, group formation
- Q3-4 2016: Use Case discussion and research
- Q1-2 2017: V0.1 Audio/Video API
- Q3-4 2017: V3.0-beta of Presentation API,
- Q1-2 2018: V3.0 for Presentation API with validator, implementations; V0.5 Audio/Video API
- Q3 2018: V0.9 Audio/Video APIs
- Q4 2018: V1.0 for Audio/Video APIs with validators, implementations
- Email: iiif-discuss, subject line: [AV]
- Slack: #av
- Github: IIIF/iiif-av
- Calls: Biweekly, Tuesday 9am PST / Noon EST: https://bluejeans.com/608386174
- Face to Face: IIIF Conference and WG meetings, others as coincidental travel allows
Chair: Jason Ronallo (NCSU Libraries)
Participants:
- Rob Sanderson (J Paul Getty Trust)
- Simeon Warner (Cornell)
- Michael Appleby (YCBA)
- Tom Crane (Digirati)
- Jon Dunn (Indiana)
- Maria Whitaker (Indiana)
- Andy Irving (BL)
- Eyal Reuven (NL Israel)
- Sheila Rabun (IIIF)
I don't see any downside to the creation of a charter. I think the proposed text is a good basis and the act of reviewing it will help focus our work. The timeline seems like a reasonable goal.
There could be a little wordsmithing here and there. The one place I think the current text is unclear is
which I think was probably intended to say:
If that is the case then I agree with the scope argument.